Case study · Production product

We built NewLedger on the infrastructure we sell

NewLedger runs real books on Paprel in production—the same ledger, API, controls, and audit model available to every Paprel customer.

Open NewLedger Inspect the architecture

NEWLEDGER

Production workspace

Live on Paprel

Illustrative net income

$84,260

Synthetic workspace data

Recent activity

balanced

Invoice paidRevenue

Bank matchedReconciled

Expense postedOperations

Report closedAugust

Ledgerbalanced

Auditattributed

Reportsready

Paprel accounting engine

1 live product: NewLedger runs on Paprel in production
3 interfaces: REST, product UI, and hosted MCP
7 years: Signed, exportable audit retention
4 statements: P&L, balance sheet, cash flow, and trial balance

Why we built this way

A customer story with nowhere to hide

NewLedger is owned by the same team that builds Paprel. Every infrastructure decision is tested against a product we must operate ourselves.

The challenge

Real books are hard to get right — and harder to embed

A production accounting product can't drift. Debits and credits have to balance atomically, every entry needs an immutable trail, reports must reconcile to the ledger, and none of it can break under real volume. Building that correctly is months of senior engineering — and most of it isn't the product's differentiator.

The approach

Build the product on the ledger, not the ledger

NewLedger runs on the same Paprel primitives we offer everyone else: a programmable double-entry ledger over REST, signed webhooks, scoped OAuth, and MCP-native tools. Product events become balanced journals; reports, exports, and audit trails read from the same source of truth.

What Paprel provides

The accounting layer underneath

Chart of accounts, journals, invoicing and bills, reporting, reconciliation, permissions, and an append-only audit history — delivered as infrastructure. The team ships accounting features instead of maintaining ledger plumbing.

The result

Production-tested in NewLedger

Paprel isn't a demo we point at a slide — it runs our own books. Every primitive NewLedger depends on is one a Paprel customer gets, tested against a real workload before it reaches you. It's the most honest proof we can offer while our first external design partners come online.

“We didn't want to sell infrastructure we wouldn't run ourselves. NewLedger's books live on Paprel in production — so every promise on this site is one we've already kept for our own accounting.”

SKSandeep Kumar · Founder, Paprel & NewLedger

Implementation architecture

One accounting path from workflow to evidence

NewLedger does not synchronize a second set of books after the fact. Its product workflows, reports, and governed agent access operate against the same Paprel system of record.

01
Capture product activity
NewLedger records sales, purchases, expenses, banking activity, and manual adjustments through controlled product workflows.
02
Compile balanced journals
The same Paprel accounting primitives turn that activity into balanced, tenant-scoped journal entries with actor attribution.
03
Read reports from the ledger
Financial statements, exports, and drilldowns resolve to the journal evidence rather than a parallel reporting store.
04
Expose governed automation
REST clients, the NewLedger product, and MCP-connected agents use the same permission and audit boundary.